The greatest impact on reducing the ever-increasing use of platelet transfusions for chronically-thrombocytopenic patients is likely to come from: 1) preventing platelet alloimmunization; and 2) improving platelet support for those patients who have already become alloimmunized. Extensive experience has been gained in a dog platelet transfusion model on techniques of preventing platelet alloimmunization. However, as thrombocytopenic patients require not only platelet but also rbc transfusions, this dog model will be expanded to include studies on how to prevent platelet alloimmunization caused by the leukocytes that contaminate rbc transfusions. The experimental approaches to be tested in this platelet and packed rbc transfusion model will include: 1) induction of immunoincompetence in transfused recipients by purine nucleoside phosphorylase mediated inhibition of T-cell function; 2) modification of the transfused blood products to reduce their immunogenicity. Modified blood products to be tested in the transfusion model will include: 1) UV-irradiation to inactivate antigen-presenting cells (APC's); 2) depletion of Ia-positive lymphocytes that include APC's by immunoadsorptive techniques; 3) production of leukocyte-poor blood products with highly-efficient filters; 4) storage or heat-inactivation of blood products; or 5) combinations of these approaches. In addition, a comprehensive evaluation of how platelet tolerance is induced by any of these approaches will involve detection of: 1) suppressor cells; 2) clonal deletion; 3) anti-idiotypic antibodies; and 4) either failure to form or loss of platelet-reactive or lymphocytoxic alloantibodies. These studies will substantially improve our understanding of methods to prevent platelet alloimmunization and why they are successful. Until these approaches to preventing platelet alloimmunization can be transferred to clinical practice, improved methods of supporting alloimmunized patients are needed. The role of platelet crossmatch testing in the support of alloimmunized patients will be evaluated in clinical trials. Two trials are planned: 1) the first will evaluate the relative merits of HLA-donor selection vs. HLA-selected crossmatch-compatible donors; and 2) the second will involve the best method of donor selection from the first trial compared to crossmatch-compatible random apheresis donors.